Quiz item analysis
Inspect binary quiz responses using item correctness proportions and correlations against scores excluding the item itself.
The example value has been entered. Check it immediately or change the value.
Input is processed in the browser. Refreshing returns a sample value.
Instructions and example
p=Σitem/n; corrected discrimination=Pearson(item,totalScore−item)
For responses 111 / 110 / 100 / 000, correctness proportions are 0.75,0.5,0.25. Each correlation uses the score with that item removed.
Supports 3–200 respondents and 2–50 items in a complete 0/1 matrix. No missing or partial scores; constant scores make correlation undefined. Not a quality verdict or ability estimate.
Frequently asked questions
Why remove the item from the total score?
It reduces the direct overlap caused by including the item in its own comparison score. The result is a sample item-rest correlation.
Does negative item correlation prove a faulty question?
No. Review sample size, answer key and content together; this tool does not prescribe automatic deletion.
Calculation criteria and reference materials
- University of Washington: Understanding Item Analyses
Context for correctness proportions and item discrimination correlations.
Verification of calculation criteria: · Calculation·verification principles · Report errors